The overall objective of this R01 grant renewal is to ascertain the pathophysiological basis of auditory/verbal hallucinations (AVHs) in patients with schizophrenia and schizoaffective disorder. The proposed research plan derives from two fMRI studies completed during the initial grant period. The first detected heightened activity prior to AVHs in the left inferior frontal gyrus (IFG) and right temporal regions that in turn was preceded by elevated functional coupling between these regions. Our second study examined resting functional connectivity (FC) summed along a functional loop linking sites in Wernicke's area, a right homologue of Wernicke's area, left IFG, and the putamen. Hallucinators with schizophrenia demonstrated robustly greater FC summed across this corticostriatal loop compared to similarly diagnosed patients without AVHs and healthy controls. These findings suggest a specific hypothesis that will be further tested, namely that heightened corticostriatal coupling linking left IFG, posterior temporal regions and the putamen leads to episodic coactivation states that are hallucinogenic. Specific Aim #1 is to utilize two speech perception tasks to elicit excessive regional activation in hallucinators compared to nonhallucinating patients and healthy controls in the same brain regions comprising the corticostriatal loop found to be hyperconnected at baseline in this patient subgroup. One task is detection of negative emotionality expressed via speech intonation, while the second task is to distinguish meaningful phrases from phrases comprised partially of non-word syllable strings. Specific Aim #2 is to use fMRI during an inner speech generation task to test the hypothesis that corticostriatal hyperconnectivity detected in our completed study is a compensation arising from pathway inefficiency. Our prediction is that evidence of pathway inefficiency in hallucinating patients will not be detected, which is consistent with our alternative hyperconnectivity->coactivation model. Specific Aim #3 is to demonstrate aberrant white matter pathways that distinguish hallucinating patients from nonhallucinating patients and healthy controls. This will be accomplished by conducting a diffusion tensor imaging study to measure fractional anisotropy in the anterior arcuate fasciculus and mean diffusivity in the inferior longitudinal fasciculus. Specific Aim #4 is to demonstrate that at rest FC linking the putamen and Wernicke's area predicts subsequent improvement in AVHs following a course of bitemporal 1-Hz repetitive transcranial magnetic stimulation (rTMS), and that level of reduction in FC induced by this intervention also correlates positively with level of improvement in AVHs obtained. Twenty-five patients with schizophrenia spectrum disorder and active AVHs, 25 patients with similar diagnoses but without a recent history of AVHs, and 25 healthy controls will be studied. These studies will provide critical new insights into the mechanistic basis of AVHs.